mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-15 04:30:12 +00:00
d4d961615c
when run on an Intel Atom processor. The failures have arisen due to changes elsewhere in the trunk over the past 8 weeks or so. These failures were not detected by the Atom buildbot because the CPU on the Atom buildbot was not being detected as an Atom CPU. The fix for this problem is in Host.cpp and X86Subtarget.cpp, but shall remain commented out until the current set of Atom test failures are fixed. Patch by Andy Zhang and Tyler Nowicki! git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@160451 91177308-0d34-0410-b5e6-96231b3b80d8
68 lines
1.9 KiB
LLVM
68 lines
1.9 KiB
LLVM
; RUN: llc < %s -march=x86 -mattr=+sse2 | FileCheck %s
|
|
|
|
; test vector shifts converted to proper SSE2 vector shifts when the shift
|
|
; amounts are the same.
|
|
|
|
; Note that x86 does have ashr
|
|
|
|
; shift1a can't use a packed shift
|
|
define void @shift1a(<2 x i64> %val, <2 x i64>* %dst) nounwind {
|
|
entry:
|
|
; CHECK: shift1a:
|
|
; CHECK: sarl
|
|
%ashr = ashr <2 x i64> %val, < i64 32, i64 32 >
|
|
store <2 x i64> %ashr, <2 x i64>* %dst
|
|
ret void
|
|
}
|
|
|
|
define void @shift2a(<4 x i32> %val, <4 x i32>* %dst) nounwind {
|
|
entry:
|
|
; CHECK: shift2a:
|
|
; CHECK: psrad $5
|
|
%ashr = ashr <4 x i32> %val, < i32 5, i32 5, i32 5, i32 5 >
|
|
store <4 x i32> %ashr, <4 x i32>* %dst
|
|
ret void
|
|
}
|
|
|
|
define void @shift2b(<4 x i32> %val, <4 x i32>* %dst, i32 %amt) nounwind {
|
|
entry:
|
|
; CHECK: shift2b:
|
|
; CHECK: movd
|
|
; CHECK: psrad
|
|
%0 = insertelement <4 x i32> undef, i32 %amt, i32 0
|
|
%1 = insertelement <4 x i32> %0, i32 %amt, i32 1
|
|
%2 = insertelement <4 x i32> %1, i32 %amt, i32 2
|
|
%3 = insertelement <4 x i32> %2, i32 %amt, i32 3
|
|
%ashr = ashr <4 x i32> %val, %3
|
|
store <4 x i32> %ashr, <4 x i32>* %dst
|
|
ret void
|
|
}
|
|
|
|
define void @shift3a(<8 x i16> %val, <8 x i16>* %dst) nounwind {
|
|
entry:
|
|
; CHECK: shift3a:
|
|
; CHECK: psraw $5
|
|
%ashr = ashr <8 x i16> %val, < i16 5, i16 5, i16 5, i16 5, i16 5, i16 5, i16 5, i16 5 >
|
|
store <8 x i16> %ashr, <8 x i16>* %dst
|
|
ret void
|
|
}
|
|
|
|
define void @shift3b(<8 x i16> %val, <8 x i16>* %dst, i16 %amt) nounwind {
|
|
entry:
|
|
; CHECK: shift3b:
|
|
; CHECK: movzwl
|
|
; CHECK: movd
|
|
; CHECK: psraw
|
|
%0 = insertelement <8 x i16> undef, i16 %amt, i32 0
|
|
%1 = insertelement <8 x i16> %0, i16 %amt, i32 1
|
|
%2 = insertelement <8 x i16> %0, i16 %amt, i32 2
|
|
%3 = insertelement <8 x i16> %0, i16 %amt, i32 3
|
|
%4 = insertelement <8 x i16> %0, i16 %amt, i32 4
|
|
%5 = insertelement <8 x i16> %0, i16 %amt, i32 5
|
|
%6 = insertelement <8 x i16> %0, i16 %amt, i32 6
|
|
%7 = insertelement <8 x i16> %0, i16 %amt, i32 7
|
|
%ashr = ashr <8 x i16> %val, %7
|
|
store <8 x i16> %ashr, <8 x i16>* %dst
|
|
ret void
|
|
}
|